|
|
|
|
|
by fusslo
738 days ago
|
|
I love these long-form build logs. I just started reading, and I am making the faux-pas of commenting before finishing. But, I'm wondering what the challenges are of automating the ROV to map a body of water's floor in a pattern. like a grid pattern, or whatever is most efficient. At first I was thinking currents would cause displacement. but can't we sense the current moving us in undesirable ways and correct with thrusters? And then I thought.. do lakes have currents? Do they have tides? can a ROV sense the boundary of a lake? just further down the rabbit hole, realizing how little I've learned about the natural environment! |
|
Autonomous accurate navigation under water is quite complicated, because after a certain point you need to start relying on local sensors because nothing reaches you anymore. But local sensors tend to be weird, because a straight line underwater is not necessarily a straight line - you are most likely drifting -- and detecting drift isn't easy. From a local observer, the water around you isn't even moving. That was a fun team to talk with.
1: https://www.iti.uni-luebeck.de/en/research-areas/mobile-robo...